Supervised Learning has to do with datasets and how the datasets are labeled. Datasets train your algorithm to classify data and determine the outcomes of your algorithm in the most accurate way possible. With supervised learning, developers can learn about the algorithm over time with increasing efficiency.
The Regression Method: this method leverages an algorithm to understand variables. These models help to predict outcomes as well. Sales revenue or LTV can be predicted using regression.
Unsupervised learning uses machine learning algorithms to analyze and cluster unlabeled data sets. These algorithms discover hidden patterns in data without the need for human intervention (hence, they are “unsupervised”).
Unsupervised learning uses machine learning to determine data sets that are unlabeled and we use them for a few main reasons:
Clustering: This is a technique used to group data sets. We label them and group them based on their differences or common properties
Association: This is a technique used to find the relationship between datasets with variables. This is often leveraged in doing basket analysis and in finding recommendations. For example, this is used to determine predictions on what customers might buy next.
Lastly, Dimensionality Reduction: This is a learning technique we use when we have too many variables in a dataset. This reduces the different variables in the dataset and keeps the integrity intact.
Ultimately, the main difference between unsupervised and supervised learning is labeling vs non labeled data sets.